Unexpected thermoelectric behavior and immiscibility of the allegedly complete solid solution $\mathrm{Sr}\left({\mathrm{Ru}}_{1–x}{\mathrm{Ti}}_x\right){\mathrm{O}}_3$

$\mathrm{Sr}\mathrm{Ru}{\mathrm{O}}_3$, a $p$-type metallic conductor with a small thermopower, and $\mathrm{Sr}\mathrm{Ti}{\mathrm{O}}_3$, a $p$-type insulator with a large thermopower in its hyperstoichiometric state, have been reported to make a complete solid solution. It is, thus, suggested that the thermoelectric power factor, conductivity times thermopower squared (σθ), may be best optimized by mixing these two extreme oxides. In order to explore this possibility, we examined σ and θ of the solution $\mathrm{Sr}\left({\mathrm{Ru}}_{1–x}{\mathrm{Ti}}_x\right){\mathrm{O}}_3$ against composition (0 $\le$ $x$ $\le$ 1), temperature (300 $\le$ T/K < 1000) and oxygen activity ($-4\le log{a}_{O_2}<0$), and subsequently its crystallographic structure against $x$ at room temperature. It has been found that, contrary to literature, there falls a miscibility gap over the composition range 0.1 < $x$ < 0.6. While there occurs a metal-to-insulator transition crossing the miscibility gap, θ remains little dependent on $T$ and $a_{O_2}$ for the compositions up to $x$ = 0.9. Unexpectedly, however, θ changes its sign twice with increasing $x$, indicating the majority carrier type changes from $p$ to $n$ at a composition in 0.6 < $x$ < 0.9 and back to $p$ at a composition in 0.9 < $x$ < 1.0, whereas σ decreases monotonically. The highest power factor is observed, contrary to the expectation, at $x$ = 0 (pure $\mathrm{Sr}\mathrm{Ru}{\mathrm{O}}_3$) on the order of 10${}^{-4}$ W/mK${}^2$. The origin of the repeated carrier-type changes is discussed in terms of electronic-structure change.

Figure 1

XRD patterns of Sr(Ru${}_{1-x}$Ti${}_x$)O${}_3$ as measured at room temperature (a) for the compositions $x$ = 0.1, 0.2, 0.3, 0.5; (b) for the compositions $x$ = 0.6, 0.9, 1.0.

Figure 2

The observed and calculated XRD patterns of Sr(Ru${}_{1-x}$Ti${}_x$)O${}_3$ for (a) $x$ = 0.6 and (b) $x$ = 0.3.

Figure 3

(a) Radial EXAFS distribution function around Ru for Sr(Ru${}_{1-x}$Ti${}_x$)O${}_3$ ($x$ = 0.0, 0.3, 0.6, and 0.9) as obtained from the Ru $K$-edge x-ray absorption spectrum. (b) Radial EXAFS distribution function around Sr for Sr(Ru${}_{1-x}$Ti${}_x$)O${}_3$ ($x$ = 0.0, 0.3, 0.6, 0.9, and 1.0) as obtained from the Sr $K$-edge x-ray absorption spectrum.

Figure 4

(a)–(d) Comparison of the radial distribution function (RDF) around Sr in SrRuO${}_3$ as obtained experimentally (red) with the RDF as calculated for different tilting angles α, β, and γ of the BO${}_6$ octahedra. (e) Crystal structure of SrRuO${}_3$ with tilting angles α = 7°, β = 6°, and γ = 2° according to Fig. 4 (best agreement between experimental and calculated RDF around Sr).

Figure 5

Linear combination fit of the EXAFS function χ${}_{0.3}$(k) of the two-phase sample Sr(Ru${}_{0.7}$Ti${}_{0.3}$)O${}_3$ with the EXAFS functions χ${}_{0.0}$(k) and χ${}_{0.6}$(k) of SrRuO${}_3$ and Sr(Ru${}_{0.4}$Ti${}_{0.6}$)O${}_3$, respectively. (a) Sr $K$ edge, (b) Ru $K$ edge.

Figure 6

(a) Electrical conductivities and (b) thermopowers vs oxygen activity of solid solutions Sr(Ru${}_{1-x}$Ti${}_x$)O${}_3$ with $x$ = 0.0, 0.6. 0.9, and 1.0 in equilibrium state at 973 K. Solid lines for $x$ = 1 are the best fitted and the small triangles show the ideal slopes: 1/4 (a) and (2.303 k/e)/4 (b) (see the text).

Figure 7

(a) Conductivity and (b) thermopower vs temperature of compositions $x$ = 0.0, 0.3, 0.6, 0.9, and 1.0. The inset in (a) shows conductivity jumps at ca. 873 K for $x$ = 0.0 and 0.3 in the linear conductivity scale.

Figure 8

(a) Conductivity and (b) thermopower vs composition $x$ of the system Sr(Ru${}_{1-x}$Ti${}_x$)O${}_3$ at different temperatures in the air atmosphere. Solid lines are for visual guidance only.

Figure 9

Positions of the Ti $K$ edge and at the Ru $K$ edge, respectively, as obtained from the XANES part of the x-ray absorption spectra for different Ti contents $x$ in Sr(Ru${}_{1-x}$Ti${}_x$)O${}_3$.

Figure 10

Power factor of solution Sr(Ru${}_{1-x}$Ti${}_x$)O${}_3$ vs oxygen activity at different compositions at 973 K (a) and vs composition at selective temperatures (b). Solid lines are for visual guidance only.